README for:

Guess, Andrew and Alexander Coppock. Does Counter-Attitudinal Information Cause Backlash? Results from Three Large Survey Experiments. British Journal of Political Science, forthcoming.

This replication archive contains the code and data necessary to replicate the analyses in the main text and in the appendicies.

It relies on a series of r packages which you can install with

install.packages(c("tidyverse", "reshape2", "estimatr", "dbarts", "stargazer", "ggpubr", "xtable"))

The archive contains 3 data files

1. GC_study_1.rdata
2. GC_study_2.rdata
3. GC_study_3.rdata

The archive contains 1 script with helper functions

1. GC_helpers.R

The archive contains 6 analysis scripts:

1. GC_study_1_analysis.r: produces Table 1, Table 2, and Table 3 in the main text and tables 5 - 10 in the Appendix
2. GC_study_2_analysis.r: produces Table 4, Table 5, and Table 6 in the main text and tables 1, 2, 11, 12 in the Appendix
3. GC_study_3_analysis.r: produces Table 7, Table 8, and Table 9 in the main text and tables 3, 4, 13, 14 in the Appendix
4. GC_figure_1.r: produces figure 1
5. GC_figure_2.r: produces figure 2
6. GC_alternative_specifications.r: produdes tables 15, 16, 17, and 18 in the Appendix


# Package version information:

> R.Version()$version.string
[1] "R version 3.5.0 (2018-04-23)"
> packageVersion("tidyverse")
[1] ‘1.2.1’
> packageVersion("reshape2")
[1] ‘1.4.3’
> packageVersion("estimatr")
[1] ‘0.9.0’
> packageVersion("dbarts")
[1] ‘0.9.3’
> packageVersion("stargazer")
[1] ‘5.2.2’
> packageVersion("ggpubr")
[1] ‘0.1.6’
> packageVersion("xtable")
[1] ‘1.8.2’

